One possible reason why the proof of concept did not work was the mirrors. Due to lack of proper equipment with the correct specs at UiO, we improvised by using filters and concave mirrors to achieve the "shortpass dichroic mirrors"-effect. Thanks to Thorlabs that were so kind to sponsor us, we got the mirrors we thought we needed. We made a change in our setup and tried to align the mirrors correctly, but as we were unsuccessful we believe that the alignment was not as perfect as it should have been.

Another possible reason why the proof of concept did not work might be that our light source did not have high enough intensity. We tried with both adding more diodes and to concentrate the light beam, but neither one of these made us see any trace of lasing with the CCD-camera. We did see some light in our spectrometer, but since it did not show up on the camera, it is difficult to say what kind of properties the light had.

Our last possible reason why we were not able to prove the concept of a biolaser is the sample of sfGFP. As we used the same sample for testing for over a month, we suspected that it began to get old. When we finally felt confident with our setup, we noticed that the sample emitted some red light it had not emitted before. Since this was only a couple of days before the wiki freeze, we did not have time to make a new sample.

In our project, we managed to create two new basic BioBrick-parts, based on codon optimized sfGFP (BBa_K2424001) and the NMT1-promoter (BBa_K2424000). We were able to clone both of these into the relevant submission vectors, and ship them off for sequencing and for the iGEM parts registry. We also managed to improve the documentation of BBa_K2110000 by sequencing and testing for the correct insertion into pSB1C3. We successfully completed the first half of our proof-of-concept by testing a protein solution containing concentrated sfGFP in our laser setup.
We were able to properly clone and submit the composite part, but due to time constraint we were unable to test our system in yeast cells. Because of this, we successfully grew S. Pombe-cells using an existing NMT1-GFP-PPK18 expressions system to grow GFP-expressing yeast cells and use that to test our laser. This, combined with the sfGFP-solutions, gave us the material we needed to be able to conduct proper experiments with our setup.

The extracted sfGFP had impurities as we can see in the results of SDS-PAGE for the first experiment. Therefore, not all 35.35 mg is sfGFP, but for our purposes, it was good enough. The main goal is to use yeast cells that are expressing sfGFP as a gain medium, and in that case as well there are impurities in the form of subcellular components of yeast cells. Therefore, that should not be a problem for our purposes.

We achieved a lot of what we set out to do, and have managed to test our setup extensively. It was unfortunate that we were able to see the light from our sample in the spectrometer, but not in the CCD-camera. Even though we aligned everything as it should be and were confident that the setup would be able to send the light into the detector on the camera, we could not see anything. Also, the fact that our sample suddenly emitted red light was peculiar, but neither our teammates or our supervisors could come up with a proper reason for why this happened.

Some tips for projects in the future is therefore to study how a proper setup should be and calculate distances and proper alignment in advance. Lastly, we believe it is wise to change the sample used for testing more frequently than we did.